INDEX

Explanations

elements that resemble structured data or identifiers, likely in a list or JSON format

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Италијани

-1.05

ArrowToggle

-0.95

 feroit

-0.91

 Italijanski

-0.90

Vidite

-0.89

 estimés

-0.88

 disambiguazione

-0.87

 auroit

-0.86

NameInMap

-0.86

ьаж

-0.85

POSITIVE LOGITS

0.62

0.61

in

0.52

0.51

 BoxFit

0.51

and

0.49

 AssemblyCompany

0.49

of

0.49

Activations Density 13.575%

elements that resemble structured data or identifiers, likely in a list or JSON format

No Comments

No Known Activations